Many attempts have been made to find a correlation between structure and activity in the treatment of anti-inflammatory diseases but no connection has been found and today researchers will agree only that arthritis is an incurable disease which defies any structure-activity relationships.
This was not always so. In 1980, I. L. Bonta.sup.1 published a review entitled "Progress in Medicinal Chemistry" where he argued convincingly that an association exists between compounds having oxygen-containing radicals as, for example, the phenothiazines, steroids, sulphydryl compounds and copper complexes, and anti-inflammatory activity. This study led to the screening of many compounds as researchers sought to find some interdependence between known anti-oxidants and the effects which those agents produce.
This anti-oxidant therory gained credence in 1985 when K. F. Swingle.sup.2 in an article entitled "Anti-Inflammatory Activity of Antioxidants" offered a rationale for using anti-oxidants to arrest and reverse the degenerative effects of arthritis in mammalian hosts. Swingle urged that since arachidonic acid, an unsaturated acid occurring naturally in fat, is known to undergo enzymatic oxidation in animals and produces pro-inflammatory prostaglandins, any compound which retards or prevents oxidation ought to be a candidate for treating arthritis and the debilitating effects of that disease.
This theory remained in vogue until about 1986 when I. Katsumi reported in the "Chemical and Pharmaceutical Bulletin".sup.3 that an in-vivo study of several di-(tert-butyl)phenols failed to support the anti-oxidant view. Support for this opinion also appears in a patent to G. Moore (U.S. Pat. No. 4,357,345) where it is stated in Column 1, lines 36-52, that the anti-oxidant activity of 3,5-di(tert-butyl)-4-hydroxytoluene, an additive used to extend the shelf-life of food, has little or no value as an anti-inflammatory agent. Moreover, the patentee notes that the absence of anti-inflammatory activity extends also to many other compounds which share the di-(tert-butyl)phenol structure as for example: 2,6-di-(tert-butyl)phenol, 4-carboxamido-2,6-di-(tert-butyl)phenol, 4-(2-chlorobenzoyl)-2,6-di-(tert-butyl)phenol, 4-(5-carboxy-2-chlorobenzoyl)-2,6-di-(tert-butyl)phenol, 2,6-di-(tert-butyl)-4-(phenylsulfonyl)phenol, 4-acetyl-2,6-(di-tert-butyl)phenol and 4-n-octyl-2,6-di-(tert-butyl)phenol. As a result, G. Moore concluded that there is no correlation which can be drawn between compounds containing the di-(tert-butyl)phenol structure and inflammation and that the effectiveness of compounds in this field can only be ascertained by trial and error (Column 1, lines 63-66).
The lack of a correlation between anti-inflammatory activity and the di-tertiary butyl phenol structure is further supported by G. Bogdonov et al. Bogdonov reports in Khimiya Geterotsiklicheskikh Soedinenii, Vol. 12: pages 1660-1664 (1971) that certain [(3,5-di-tert-butyl-4-hydroxyphenyl)alkenyl]pyridines, specifically, 2-[2-(3',5'-di-tert-butyl-4'-hydroxyphenyl)ethenyl]pyridine and 4-[2-(3',5'-di-tert-butyl-4'-hydroxyphenyl)ethenyl]pyridine, are effective in the treatment of tumors; however, the authors fail to attribute to these compounds any evidence of anti-inflammatory effectiveness.
The literature, however, is not without reference to 2,6-di-tert-butylphenols bonded directly to a heterocyclic ring as shown, for example, in European Appln. No. 59,090. Moreover, Lazer in U.S. Pat. No. 4,743,606; Moore in U.S. Pat. Nos. 4,172,082 and 4,124,725 and Isomuda et al in Chem. Pharm. Bull. Vol 31: pages 3179-3185 (1983) describe a variety 2,6-di-tert-butylphenols bonded directly or indirectly to pyridine and other heterocyclic or aromatic nuclei; however, none of these are pyrrolidinones, piperidinones or functionally equivalent heterocycles bonded directly to a 2,6-di-tert-butylphenol moiety.